庫力索法 (KLIC) 2022 Q4 法說會逐字稿

Greetings, and welcome to the Kulicke & Soffa 2022 Fourth Quarter fiscal results call. (Operator Instructions) As a reminder, this conference is being recorded. It is now my pleasure to introduce your host, Joseph Elgindy, Senior Director of Investor Relations for Kulicke & Soffa. Joseph, you may begin.

Thank you. Welcome, everyone, to Kulicke & Soffa's Fiscal Fourth Quarter 2022 Conference Call. Fusen Chen, President and Chief Executive Officer; and Lester Wong, Chief Financial Officer, are both also joining on today's call. For those who have not received the recent results, the earnings release as well as our supplemental earnings presentation are both available in the Investor Relations section of our website at investor.kns.com. In addition to historical statements, today's remarks will contain statements relating to future events and our future results.

These statements are forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995. Our actual results and financial condition may differ materially from what is indicated in those forward-looking statements. For a complete discussion of the risks associated with Kulicke & Soffa that could affect our future results and financial condition, please refer to our recent SEC filings, specifically the 10-K for the year ended October 2, 2021 and the 8-K filed yesterday. With that said, I would now like to turn the call over to Fusen Chen for the business overview. Please go ahead, Fusen.

Thank you, Joe. Over the past 5 years, we have evolved into a more resilient, growth-oriented and a dramatically more profitable company by focusing for our corporate cultures, strengthening our established positions and expanding our served available market. In parallel to this fundamental internal effort, semiconductor assembly, we did both the high volume and leading-edge market is now a more significant contributor to the industry's value chain. Over the past year, we flexed our capacity and overcame broad supply chain disruption to support customers through a rapid period of industry expansion. In parallel, we executed on multiple advanced development projects and continued our market expansion strategy.

Collectively, this ever has increased our base level of revenue and are clearly represented in our financial results. Through fiscal 2022, we again generated over $1.5 billion of revenue, in line with the fiscal 2021, although our non-GAAP earnings per share increased by 21% over the same period. This higher level of performance increases our resiliency as we look into fiscal 2023. Over recent months, industry leaders and the forecasters lowered WFE and the semiconductor unit outlooks due to the increased uncertainty related to interest rates, global trade tension and ongoing supply-chain disruption, which negatively impacted both inventory and the demand levels across general semiconductor, LED and the memory end market.

Considering this dynamic environment, we recently conducted scenario planning across our individual business lines. Based on this detailed feedback, we currently expect fiscal 2022 revenue to meet or exceed our previous cyclical peak revenue in fiscal 2018. This outlook suggests a more typical seasonal pattern through fiscal 2023, with ongoing digestion in the first fiscal half, followed by a gradual demand improvements in the second fiscal half. The expected second half improvements are supported by well-known seasonal dynamics in addition to a heavier weighting of advanced display and advanced packaging revenue.

Despite this dynamic macro and industry environment, secular trends in advanced display, advanced packaging and automotive has continued to be very resilient. Lester will provide additional detail on our outlook shortly. Over the prior years, we generated revenue of $1.5 billion and a non-GAAP EPS of $7.45, representing an increase of more than 2x over our prior 2018 peak full year, which helped to highlight how our cyclical performance has improved. While semiconductor growth has contributed, since 2018 prudent partnership, acquisitions and aggressive development expanded our served available market by 51% to approximately $4.7 billion.

This change, which excludes our pending acquisition, provides a more sustainable and consistent path for growth going forward. We remain focused on our long-term strategies and the outlook over the coming years. Fiscal 2023 is a critical adoption period for our higher growth solution supporting advanced packaging, automotive and advanced display, which are increasingly aligned with the long-term fundamental technology transition already underway. Despite the softer environment, customer engagement and the interest for our growing portfolio of solutions continue to expand.

I will provide an update to this key growth initiatives shortly. In addition to our ongoing organic development effort, we have been seeking competency-based acquisition that can further accelerate our growth potential. On September 8, we announced an agreement to acquire Advanced Jet Automation, AJA. AJA's technology portfolio will further expand our served available market while also materially increasing our access to the evolving micro and the mini-LED opportunity.

Over the past three years, success with Assembleon acquisition has allowed us to enter the advanced display market. This access offered the opportunity to identify and engage with the innovative providers, including AJA, who are also supporting this emerging high growth opportunity. AJA's unique and complementary dispense solution which provides market-leading placement accuracy and repeatability -- already addressed the high accuracy needs of Advanced Display and are positioned to address the growing complexity of semiconductor and the consumer electronics assembly.

In total, AJA provides K&S with access of roughly $2 billion total addressable market in dispense, providing an additional layer of growth. This sizeable new market access and impressive competencies in the emerging advanced display space supplement our broad organic growth initiative. Our existing technical competencies, sales and distribution network and the operational strength can help AJA better commercialize new solutions and accelerate growth potential.

Turning to our end market, we generated $242.1 million of sales from our capital equipment businesses in the September quarter, which represents a 23% increase over our 5 years' average. Within the general semiconductor market, a softer outlook is expected due to the lower consumer spending level and also indirect effects of new trade restrictions. However, we continue to execute on share gains in the power semi market, advanced packaging and soon also within electronics assembly.

Our Advanced Display business is progressing better than expected, and we significantly exceeded our $80 million advanced display revenue target. Advanced Display represents nearly 60% of our total '22 LED revenue. As we execute on development and further drive adoption across a growing customer base, we expect this market to grow materially. After several quarters of rapid automotive capacity expansion, we have returned to a more reasonable level of demand for our core-automotive solution. We continue to closely support our automotive customers through our leading semiconductor, electronics and battery assembly solutions, which are directly addressing many (inaudible), power management, power storage and the power distribution needs for current and future electric and autonomous vehicles.

These trends are significant and expected to continue supporting above average automotive semiconductor growth over the long term. Today, we continue to expand our fundamental strength by supporting our own capacity expansion plans and the new product initiatives while delivering on several intimate customer engagement. Our ongoing progress and execution increase optimism into FY '24. Allow me to provide a brief update.

First, we now have multiple facility expansion and the renovation program in Singapore and Pennsylvania, which are providing critically needed clean-room lab and the metrology space that will enhance our manufacturing and development capabilities. These expansion efforts better support the growing trend and the demand for our new advanced packaging and advanced display solution. Our dedicated semiconductor advanced packaging business has grown by 34% over fiscal 2021 and is projected to continue growing materially over the coming years.

Customer interest and feedback for our fluxless thermocompression system have increased over the past quarters, and we continue to expect this process will address the majority of heterogeneous assembly needs down to a 10-micron pitch. We currently have an industry leadership position in chip to substrate process and have multiple promising new opportunities with key customers in chip-to-wafer process over the coming years. In addition to our focus on emerging heterogeneous integration opportunity, TCB also support the high growth, high-volume, system-in-package market for emerging logic processor, mixed signal, silicon photonics and sensing applications. This new access to high growth opportunity provides a specific example of how we expanded our market reach and are raising our base level of business.

As the value of semiconductor assembly increases, our engagements with multiple fabless companies have also increased. While these are not traditional end customers, the assembly process is clearly becoming a more significant factor in IC design than in the past. Demand for our new solutions continues to improve across our growing base of fabless, foundry, IDM and OSAT customers, and we are working aggressively to support broadening customer engagements.

Considering this new momentum, we anticipate thermocompression to provide meaningful growth over the coming years. To highlight this momentum, our thermocompression businesses grew by nearly 5x year-over-year. We have also currently identified specific TCB customer opportunities of over $300 million, cumulatively, through 2025. In addition to advanced packaging, we are strategically focused on expanding market share through the pending release of our latest electronic assembly system.

Looking back, our 2015 acquisition of Assembleon provides several market expanding opportunities for K&S, including additional access into the automotive markets, new access into system-in-packaging flip-chip market and also new access into the emerging mini- and micro-LED space through the success of PIXALUX. After securing position in these adjacent markets, we are also targeting share gains within the core electronics assembly market. Recent and ongoing development efforts have positioned us well to expand our access within electronics assembly, which represents a served available market in excess of $2 billion.

Over the past year, we have delivered a new system architectures, which addresses the growing accuracy and the throughput needs of next-generation electronics assembly. Initial customer feedback has been well received, and we look forward to officially releasing our latest system in the second half of fiscal 2023.

The last update is regarding our growing portfolio of advanced display solutions, which continues to track to expectations into fiscal 2023. Sustained development efforts have created multiple advanced display solutions, PIXALUX, LUMINEX and also close customer development programs, which comprehensively address the LED placement requirements of emerging backlighting and direct-emissive applications. At a level -- at a high level, LED technology which represents the vast majority of display production will benefit significantly from emerging backlighting trends over the long term. To be clear, LCD technology offers a lower production cost and the lower useful life than current alternative display technology like OLED. Emerging backlighting trends supported by success of PIXALUX and the growing interest of LUMINEX further optimize the cost over performance tradeoff for LCD technology, specifically with a larger format displays.

Alternatively, OLED technology provides a thinner, higher quality image supporting high relative shares with a smaller format display, although image degradation and the production costs have limited broad OLED adoption in high-volume, larger format display market. Over the coming years, direct-emissive displays, using only a dense matrix of very small LEDs have the potential to challenge OLED technology from a performance and power efficiency standpoint. This trend is only beginning to play out, and we are well positioned to participate through our close customer development initiatives and also the success of LUMINEX over the coming quarters.

For both advanced backlighting and direct-emissive approaches to be adopted, lowering production costs is critically important to driving market adoption while production costs of mini and micro LED will improve, we are most focused on delivering higher throughput solutions, while support both of these long-term trends. Our first-mover position and success with PIXALUX has provided us with the largest installed base of ultra-high-speed pick and place tools for advanced display. This level of performance improves with LUMINEX laser-based transfer method. With LUMINEX, we are on track to achieve 3x the productivity benefits of PIXALUX over the coming months. Our R&D teams are actively supporting several ongoing qualifications in parallel for LUMINEX as they reach this new milestone.

We continue to expand our advanced display installed base and pursue multiple LUMINEX engagements while making consistent progress across several customer development initiatives. Customer interest in our latest LUMINEX system remains very strong. We have also recently shipped a new customer-specific advanced display solution that further increase our optimism and the long-term potential with these broad technology trend. While near-term industry growth rates routinely change across the semiconductor market, our positions have fundamentally improved and better correlate with a secular trends, shaping the future of advanced packaging, automotive and the advanced display markets.

Through ongoing execution of our development programs, integration of AJA and driving customer adoption, we are well positioned to further enhance our market access, growth prospect and the fundamental strengths over the near term. With that said, I will now hand the call over to Lester, who will discuss our financial performance and the outlook. Lester?

Thank you, Fusen. My remarks today will refer to GAAP results unless noted. First, I would like to address the recent changes to U.S. trade regulations, which have clearly impacted many finance solution providers' ability to support existing production and ship tools to many Chinese companies involved in IC fabrication. While we are still receiving confirmation from customers, we do not anticipate any material direct impact to demand. The vast majority of the systems we ship into China are simply not restricted. Additionally, none of our products support IC fabrication. Our products only support IC assembly, which is excluded from the new restrictions. While we don't anticipate direct impact, the new rules will likely create near-term supply chain disruptions, which may indirectly impact demand for our products.

As Fusen explained in detail, the current macro environment remains dynamic, and we remain committed to expanding our product portfolio and market access in a fundamental, long-term and sustainable way. Our strategic path includes many facets, customer engagements, technology partnerships, development programs, qualifications and highly selected acquisitions, which sustainably enhance our technology-orientated growth. As we strategically expand our market access and product diversification, through cycle operating leverage and free cash flow generation will continue to improve.

Over the past fiscal year, we quickly flexed our manufacturing to meet an unprecedented level of demand for our high-volume ball bonder business, generating revenues of over $1.5 billion, non-GAAP net income of $455.6 million and non-GAAP earnings per share of $7.45. As Fusen mentioned, non-GAAP EPS actually increased by 21% year-over-year despite a similar level of revenue.

The largest individual driver to this benefit was due to our 380 basis points improvement in gross margin during fiscal 2022. To be clear, the growing capital intensity of the semiconductor assembly process is directly benefiting the value proposition and long-term growth rates of our leading ball bonding solutions. Our strengthening position in advanced display, advanced packaging, automotive and electronics assembly may be easier for investors to digest, although we have also optimized the core high-volume ball bonder business. While ball bonding has historically been underappreciated, the underlying business has fundamentally improved.

Since fiscal 2020, our ball bonder gross margins have increased by 340 basis points, largely due to stronger demand for our high-performance systems, which are more capable to run multi-die applications. Specifically, the rapid series of ball bonders, our most advanced architecture grew from representing only 21% of total ball bonder units in fiscal 2020 to representing 69% of total ball bonder units in fiscal 2022. New complex packaging trends enhance our existing leadership position and the longer-term growth rates within this large, well-established process.

Turning to our recent results. During the September quarter, we generated revenue of $286.3 million, gross margins above 46%, non-GAAP net income of $70.2 million and non-GAAP EPS of $1.19. Gross margins came in slightly below our guidance range, largely due to accounting associated with customer-related development efforts. Non-GAAP operating expenses during the quarter came in better than expectations due to immediate cost control efforts, a reduced pace of hiring and foreign exchange gains related to a strengthening U.S. dollar. Finally, tax expense for the quarter came in at $6.6 million, slightly better than expectations.

Over the past year, our total cash position increased by $35.7 million after committing $322.2 million to investors through dividends and share repurchase activities. Working capital days increased to 341 days in the September quarter, representing a sequential reduction in revenue, sequential increase in cash and a collective decline in accounts receivables, inventories and accounts payable. Through fiscal 2022, we generated $367.4 million of adjusted free cash flow, highlighting our longer-term earnings potential.

During the September quarter, we repurchased an additional $60.2 million of shares, bringing our fiscal year total to $282.8 million, which represents 5.6 million shares or nearly 10% of our fiscal 2022 weighted diluted share average. At the end of the September quarter, we had nearly $250 million remaining under our repurchase authorization and continue to manage an active open market repurchase strategy. In addition to the repurchase activity, we have also just announced our third consecutive annual dividend raise, bringing our total dividends per share to $0.76 annually and maintaining a competitive dividend yield. The ongoing repurchase program and steadily growing dividend help stabilize our valuation while optimizing our fundamental market expansion effort on a per share basis.

For the December quarter, in line with Fusen's comments regarding softening macro and industry environment, we anticipate revenue of approximately $175 million, plus or minus $20 million. Gross margins are expected to reduce to 45%, plus or minus 50 basis points due to product mix, accounting related to our customer-developed initiatives, additional expediting fees, near-term facility resizing efforts and also higher-than-expected inflation.

Non-GAAP operating expenses is anticipated to be approximately $68 million, plus or minus 2% due to ongoing expansion efforts in addition to inflation. Over the last month, we have reduced our rate of hiring and limited noncritical expenses as we have during prior soft demand periods. Non-GAAP EPS is expected to be $0.20 plus or minus 10%, which considers an effective tax rate of just over 20%. This increase is partially related to regional income mix, although it's primarily due to mandatory capitalization of R&D expenses under Section 174 beginning in fiscal 2023. Unless repealed or modified, this provision of the Tax Cuts and Jobs Act of 2017 is expected to broadly affect all U.S. corporate taxpayers with R&D activities.

As we look further through fiscal 2023, we continue to anticipate a period of capacity digestion to extend into the March quarter with typical seasonality trends driving more distinct capacity needs in the second fiscal half. It remains a very exciting time for the company as we have been significantly broadening our alignment with fundamental technology change across the semiconductor, advanced display, electronics assembly and automotive markets.

As the industry recovers and we continue to execute, we are well positioned to reach new levels of financial performance beyond 2023. While there are near-term challenges for the entire industry, our fundamental improvements, enviable financial position and active roles enabling several long-term technology transitions will allow us to emerge as an even stronger and more profitable company.

This concludes our prepared comments. Operator, please open the call for questions.

(Operator Instructions) Our first questions come from the line of Krish Sankar with Cowen.

I had a few of them. First one, Fusen, when you look into the December quarter and subsequently into the March, is it fair to assume pretty much all segments are sequentially down in December and in March, like general semi memory R2 and LED? Or is there a trend in any of them into the December quarter?

Well, I think we are seeing our revenue in the perspective of the previous quarters. We feel other than the unit growth related products, actually, other product hold pretty well. So we feel like Q1 and Q2 are probably (inaudible) and we could expect situation probably get better after that. I think -- I wish I answered your question.

Got it. And then just out of curiosity, what kind of gives you the comfort that your fiscal second half demand will recover, i.e. from the June quarter onwards?

So there are a few things. One is, of course, our customer feedback. We continue, however, to talk to the customer. I think inventory digestion (inaudible) maybe they are more focused actually by the second half. And also, from our point of view, I think we have advanced display and advanced packaging, they are more weighted in the second half. And also, I think a few market forecasts of your (inaudible) way, probably, I think after March quarter the situation will get better.

Got it. Got it. All right. And then I just had 2 other quick questions. In the September quarter, what was your percentage of sales to China?

It's just around 50%.

50%. So it's kind of been in the same range for a while though now, right?

Yes. But 20% of that actually is not -- it's the international customers, but with their factories in China.

Got it. And then a final question. In the past, I think you kind of mentioned that in FY '22, advanced display is about 60% of total LED sales. Do you expect that dollar value to decline in FY '23 for advanced display given that your LED revenues are right now? Or do you actually think year-over-year FY '23 advanced display revenues would grow?

Well, we anticipate that advanced display for FY '23 will be higher, about similar to FY '22, roughly about $100 million.

So I think in my script, I mentioned all authorities this way. Actually, in (inaudible) you'll continue this further out with enlarged area of -- opened area of our projects. So that's the PIXALUX, LUMINEX, we are actually in the market (inaudible). And we also (inaudible) shipped product -- shipped customer product and all these are associated together. Actually, the downturn also impacted a little bit of this advanced display. But because it is (inaudible) $80 million to $100 million.

Our next questions come from the line of David Duley with Steelhead Securities.

I guess, first of all, you talked a lot about the thermal compression bonding opportunity. I was wondering if you could just elaborate a little bit more on what you think the size of that market is in dollars on an annual basis? And what your market share is? And then your competitor, I think, on their conference call was making a big deal about working with AMD or their logic provider. Could you talk about who your key customers are? What end markets your efforts are in?

Okay. So I think TCB in the recent -- just recently, I think the prospect actually increased a lot. (inaudible), I think also people transferred as advanced free chip to advanced thermal compression. So I think in my previous call, we mentioned we have a backlog of $18 million. So of course, we ship to our customer. In the meantime, we also get a new [PO]. But (inaudible) PO, I think we have a requirement. It's not only order. We're going to have a specific delivery. So sometimes I think our backlog is not the better judgment. We actually conduct a study after we visit a lot of customers from (inaudible) design house, (inaudible) and many, many customers. We feel that actually, this is a very actually promising market.

So let me make a clarification. I think TCB, particularly, we're talking about Fluxless TCB. We have a Fluxless because of -- when the TCB process enter into above [30] micron, the flux become a contamination. We also have a proprietary special technology. We are able to make a very clean silicon to silicon content, so when -- actually, the biggest volume at this moment, actually, is between 30 to 10 microns. So our Fluxless TCB differ from hybrid bonding in 2 ways.

Number one, I think hybrid bonding is focused on pitch below 10 microns. And I continue to say we are focused on up to 10 microns. So actually, our focus around at this moment is between 30 to 10 microns. And actually, at this moment, it's a very large market, right? So that's the first difference is, at this moment, these 2 technologies are not overlapped. The second difference actually is our fluxless TCB is a pure back-end process, but high rebounding is assisted with some planning process. So some China investment is needed, right?

So at this moment, it's really not overlapped at all. We don't know if any later, they will have overlap. At this moment, we did not claim to take any market from (inaudible). So within the TCB, actually, are still dealing with two processes. One is chip to wafer once the chip will substrate. We actually are the leader for chip to substrate. And we actually are shipping a new chip-to-wafer system to our significant customers. And we do expect that we will gain market shares for the chip-to-wafer process, and at this moment, I think chip-to-wafer and chip substrate, they are equally large, right? So I hope I answered your questions.

Yes. Just as a follow-on to the thermal compression bonding. This is a process that's used mainly with -- in the Logic segment, right, as your competitor has talked about it. Are there other mark -- is that where you're seeing your successes with Logic providers? And are there other markets that would be adopting this technology?

Well, you mean thermal compression, right? Yes, I think high bonding with memory, of course, people also talking about hybrid bonding. We don't have a comment on that. So let me go back. I think we mentioned about identifying high potential over $300 million. Previous call, actually, we said $80 million backlog. This pretty much will be shipped within '23 and the $300 million we talked, the majority we will ship in 2025.

Okay. And then I just wanted to clarify, in your prepared remarks, I think you said something comparing your current business levels to 2018. Could you just repeat what you said? I just didn't hear what you were trying to...

So let me make this comment. I think during our business cycle, the upturn actually is down from 2020 to 2021. So our 2020 revenue is [630], I think, if I really go right. And we ended 2021 with [$1.5 million]. So actually, the growth rate actually is [168%]. So I think in the downturn, after the second trade restriction to China, actually, a lot of people have [died down] with set expansion about roughly 15% to 20%. So this will also indirectly impact the demand for the back end. So we are part of that. And in addition, I think the unit growth rate also revised down. So therefore, it's very difficult to predict how the annual of FY '23 is going to be ended.

But after detailed, actually, study internally, we feel that we comfortably can meet the 2018, which was the previous cycle peak at around $900 million, actually more precise is $890 million. We feel like this is attributable from our side, and it's quite difficult. And we feel like we can do at least or better than that. It's quite difficult to (inaudible) for you. So that's what I -- so we feel like we are at probably $900 million, probably will be about like high 30 some, (inaudible) but the upturn actually brought us about 168% up. And the worst-case scenario, we are seeing probably downturn -- well, pretty much down above (inaudible).

Okay. Final question for me. Lester, you mentioned how you've seen improvement in the core wire bonder gross margin over the last few years, I think it was 340 basis points. So a very robust number. But correct me if I'm wrong, don't you have a new wire bonder coming out next calendar year that should also help improve gross margins? Could you elaborate a little bit more on that?

Yes. We continue to introduce new products into our core business. So we believe that the ball bonder gross margin will continue to remain high, and we'll continue to look for ways to increase the margin on our core business in ball bonder as well as wedge bonder.

So David, in downturn, actually the demand and the market share and the price also has some correlation. So we will do the best to increase the gross margin while we also make sure we get enough market share.

Yes. I guess what I was referring to is in the past, you've talked about being able to improve overall margins by 400 or 500 basis points, and I think that was kind of from a 47% level. And I realize you're going into a downturn. So during that period, gross margin improvement is -- doesn't happen. But when we get back to, I guess, normal levels at 1 point or another, do you still think you can improve the overall gross margins to that level?

Yes. Our target has always been 50% gross margins in a non-downturn year, right? And we believe that in '24 and beyond, we can reach that goal, if not higher.

Our next questions come from the line of Craig Ellis with B. Riley.

Team, congratulations on the dividend increase and the cash used to share buybacks and some check focused M&A. So Fusen, I wanted to start just by seeing if you could provide some color around some of the fiscal '23 commentary. So very helpful to hear that it seems reasonable from the company's view that sales might be down 30% to 40%. The question is, as you look at that and going back to the comments about customer interaction and conviction that they have that things can move up in the back half of the year, how should we look at trends for general semi advanced display, auto and industrial and memory? Which of those would be relatively stronger next year? Which relatively weaker based on what you're hearing from your customers?

Well, I think memory probably is looking at the '24, but the later part of '23, I think it might have some chance. But memory, I think everybody knows because at this moment is a little bit weak. For the display, actually in the consumer market, unless I think that's why we also need to focus on our high throughput. The display actually is also not a very easy market. So for the consumer part it is going to be impacted a little bit. But for the long term, I think we are a firmly believer that mini LED and the micro LED is here to stay. And we also have a lot of qualification activity with customers. And so it's really not only like volume related, there is a lot of the new projects.

So in terms of our advanced display, we are looking at 2021, I think our revenue was $80 million. 2022, actually last quarter, we ended last quarter at $90 million. But after this quarter, the full year, I think we are slightly over $100 million for advanced display. For '23, we actually have a target to be $80 million to $100 million advanced display. So advanced display, I think is a little bit of weakness. But we have a lot of qualification. Customers use this opportunity to qualify the products.

So advanced packaging, I think we actually cannot ship more than enough. And a lot of request, actually, we are increasing our capacity and (inaudible) customer's demand, right? So I hope I answered your question. And for the unit growth of related products, actually, we have 2 (inaudible) go longer. We feel like actually already significantly -- the inventory actually issue -- actually I think at this moment, we are at a quite low level already. But another 1 actually is wedge bonder. Wedge bonder, I think in our 2018, only qualified with our first EV customer, the revenue is $100 million. And this quarter, actually, we are looking very close to (inaudible). So this is riding the auto trend, so which I give you some color about our customers.

Yes. That's really helpful, Fusen. Lester, I wanted to understand more about what was happening with operating expense control. I think you talked about slowing hiring and a few other things that happened tactically. And then there was some, I think, FX benefit. So can you quantify the FX benefit and what should we expect with operating expense quarter-on-quarter. I'm sure we're down just given variable cost model, but are there incremental, tactical or structural cost savings that are coming into the model as we look at fiscal 1Q?

Yes. So thanks, Craig. I think in Q4, we did have about $4 million positive ForEx that helped to bring the OpEx down. I think the other thing is we are intimating cost control, but as we have always done, we focus on our critical projects. We continue to invest in our critical projects. And I think in Q4, we budgeted in hiring certain personnel in critical projects and R&D. The labor market is still a little bit tight, so some of those hires did not happen. We expect that to happen in Q1. So of course, we will continue to look at the noncritical controllable costs. We'll push it out to the second half, if we can or delay it all the way to FY '24. But for critical projects in advanced display and advanced packaging, we will continue to invest in Q1 and throughout FY '23. We believe that, that will put us in a very strong position. When the recovery comes in '24, to really ramp and take advantage of that.

That's helpful. And then lastly, [Vivek] talked about some positives that you're seeing in the compound semi part of the business, and I was hoping you could just elaborate on what you're seeing and what investors could expect in fiscal '23?

I'm sorry, I probably is misunderstanding. Are we talking about high power semiconductors, right? So that really is always...

So is that IGBTs or are you talking about silicon carbide?

No, IGBT. Yes.

Okay. And so I think if that's auto related?

I'm sorry? Yes. That's right.

Is that an auto related application?

Yes. That's correct.

Our next questions come from the line of Charles Shi with Needham & Company.

I want to go back to your comment on fiscal '23 revenue number. You think it's going to be closer, I mean exceed the fiscal '18 number, roughly $900 million. You already guided the first quarter, roughly $175 million. And you also provided that in March quarter, you expect a little bit more like a bottom level run rate quarter, but you expect the second half to make up more of the growth to get you to roughly $900 million for the full year. But that would require a second half June quarter -- September quarter run rate to be somewhere closer to $300 million per quarter. How do we bridge between like your guidance for the first quarter fiscal and the second fiscal quarter something like well below $200 million to something like close to $300 million in the second half? That's my first question.

Charles, thanks for the question. So traditionally, the second half has been much stronger than the first half. In fact, historically, it's been a 60% plus of the year's revenue is in the second half. And Fusen did say that, yes, the first half is softer. The trough will either be Q1 probably and Q2 maybe up a little bit, a little bit flat. But we also, as we indicated, based on the macroeconomic factors which we talked about, right, in terms of -- within semiconductor or also just general macroeconomic factors improving in the second half of our fiscal '23, we have, again, some of the new projects. We'll start providing more meaningful revenue in the second half of '23. So I think between all those factors, we do believe that the second half will be stronger than the first half, and it does provide us a path to -- for FY '22 to exceed FY 2018.

Yes. But you are basically assuming -- I think you mentioned about capacity digestion that's going to only last a couple of quarters for you this down cycle. But if I look at historically, at least the '18/'19 cycle, yes, the digestion probably actually lasted about 2 years. How do you think this cycle is going to be different from the last down cycle, if you're assuming like a relatively weak capacity digestion here?

Yes. So Charles, I think we are not saying it's going to be at a very, very high level. So at this moment, if you look at our -- the evenly related product, I think it's probably very, very low already. So if we have a 2 quarter of this similar towards 360 -- so about 360, right? So let's make a comment. I think -- historically, I think our second half, it's about 60%, right? So 60% comes, say, at $900 million, that's [540]. So if you add this up, actually I'm quite lucky about that $900 million already. And we also feel like we probably have a strength on some products integrated in the second half.

And even $300 million compared to actually upturn, it was a quarter, I think the peak we reached over $480 million. So I think this is a very strange cycle. It's because about 2019 was established with trade engine. And when we start to expect to go up, there is pandemic, and then a lot of things actually, inventory higher up in China. So it's very difficult to explain, I think, our business cycle, but we do believe about $300 million per quarter. I think it is not super difficult for us to achieve just a little bit pickup of the unit orientated products. I think we should be able to achieve that. And coupling with the advances, there is very few products we have momentum, I think it's achievable for us for the 2018 high. That's our opinion.

Got it. Got it. So you have stopped disclosing quarterly backlog. But I think you're still obligated to disclose your annual backlog number. Since this is your fiscal year-end, can you provide what the number is?

$531 million.

How much again? Sorry.

$531 million.

Got it. Got it. So next question, you talked about AJA acquisition. How much of the annualized revenue run rate is that business? Can you give us some number there?

Well, Charles, I think, obviously, for FY '23, we're integrating the business, right? And so I think as we move forward in FY '24 and beyond, we think there's significant growth given the size of the Dispense market, right? But I think for FY '23, we're looking at probably a little bit north of $10 million for AJA, but we will not have AJA for the entire fiscal year. As Fusen mentioned, we'll probably do only have it for the second half.

Got it. Got it. So maybe for the sake of time, my last question, I really want to ask you about OpEx. You guided -- well, first off, your September quarter non-GAAP OpEx is somewhere about $59 million, if my math is right. But you're guiding December quarter non-GAAP OpEx $68 million. I think that's still a big amount of uptick from the September quarter level. Given the macro environment, should we think about a little bit more cost control than that what your guidance implies?

Well, Charles, we do have very stringent cost control as I already mentioned, right? For noncritical controllable interest -- sorry, expenses we're watching very carefully as we did in the previous soft quarter, and we'll continue to do so. But as I indicated in my earlier remarks or answer to a question, we will continue to invest in the critical projects, particularly in advanced packaging, advanced display, electronic assembly as well as our core business because we believe that those are very exciting opportunities.

I think Fusen already mentioned this -- what the positions that we believe we can take in both advanced display and advanced packaging in '24 and beyond. And also when the recovery comes back, I think with the investments in our core business, we'd be able to increase margins as I responded to Dave as well as gain additional market share. So we are very careful on cost control, but we also understand that you need to invest in order to be able to grow the business in the future, and that's what our philosophy always has been.

Our next questions come from the line of Hans Chung with D.A. Davidson.

I have a couple. First, what's the underlying assumption for semi unit growth for fiscal '23? Given your commentary on '23, I say something nearly -- like nearly $900 million or both. What is the assumption for semi unit growth for that?

We assume semi unit growth for FY '23 to be about flat to plus or minus 2%.

Got it. Got it. And then I think so far on that -- so given that we think you can do $900 million level top line for '23, and it seems like we also continue to invest in '24 and beyond. So since -- from a bottom line perspective, the -- I guess the [EPS] number will be probably those in the level in 2018? Or any color you can provide regarding the bottom line for fiscal '23?

Well, Hans, we don't guide beyond the quarter, right? But I think as far as the EPS or the GAAP net income at least for FY '23, we provided we think we can do revenue better than '18. I think the gross margin for the year is probably going to be around 46% to 48% increase, the better as we move into the second half of the year. And I think we've sort of given an indication what the OpEx number should be. So I think based on that, and then we'll also provide some color on that in terms of tax. So I think based on those, I think your model should be able to generate what you think the EPS would be.

Okay. Got it. And then -- and lastly, so I think last time, we kind of talked about the SMT opportunity. And I just wonder if there is any update. And then I think since that you are -- you were talking about like the share gain story on the new generation tool. And then just wondering like what kind of the competitive advantage they allow you to -- is that -- would that be something like the second half '23 story or more like in '24?

Yes. Actually, this is a new SMT system, so it's our electronics assembly. There are a lot of competitors over here. We do believe the new innovation is -- the head -- is very, very fast, right, because (inaudible) hit and we are going to officially release probably second half of 2023. So in terms of revenue impact, probably we are looking at probably 2024, right? But we actually have a quite confidence compared to all the existing leaders, our throughput and reliability, I think, will be very, very competitive. So that's the update. We actually tested the market and have a very good feedback, but we can only officially release second half of 2023.

There are no further questions at this time. I would now like to turn the call back over to Joe Elgindy for any closing comments.

Thank you, Darryl, and thank you all for joining today's call. Over the coming months, we will be presenting at several investor conferences hosted by Needham, the Susquehanna Financial Group, in addition to the Annual New York City Summit. As always, please feel free to follow up directly with any additional questions. This concludes today's call. Have a great day, everyone.